Building the Hierarchy Bridge Table

By Ralph Kimball

In our books and classes, we describe a powerful technique for representing complex ragged
hierarchies of indeterminate depth. The centerpiece of the design is a bridge table whose grain is every
path from any given node in the tree to all the descendants below that node.

Please see pages 215 to 224 in The Data Warehouse Toolkit, Third Edition, for an in-depth discussion
of this approach to modeling ragged hierarchies.

At first glance it is not obvious how to build the bridge table. Typically, the original representation of
the tree structure is a set of recursive records, where each record represents a node in the tree and
contains a recursive pointer to the unique parent of that node. For example, if we had a fictitious
company called Microsoft whose organization structure is represented by 18 nodes, the original
representation of the tree could be stored in a Company table as follows:

Create table COMPANY (

COMPANY_KEY INTEGER NOT NULL,
COMPANY_NAME VARCHAR(50),
PARENT_KEY INTEGER);

INSERT INTO COMPANY VALUES (100,'MICROSOFT',NULL);

INSERT INTO COMPANY VALUES (101,'SOFTWARE',100);
INSERT INTO COMPANY VALUES (102,'CONSULTING',101);
INSERT INTO COMPANY VALUES (103,'PRODUCTS',101);
INSERT INTO COMPANY VALUES (104,'OFFICE',103);
INSERT INTO COMPANY VALUES (105,'VISIO',104);
INSERT INTO COMPANY VALUES (106,'VISIO EUROPE',105);
INSERT INTO COMPANY VALUES (107,'BACK OFFICE',103);
INSERT INTO COMPANY VALUES (108,'SQL SERVER',107);
INSERT INTO COMPANY VALUES (109,'OLAP SERVICES',108);
INSERT INTO COMPANY VALUES (110,'DTS',108);
INSERT INTO COMPANY VALUES (111,'REPOSITORY',108);
INSERT INTO COMPANY VALUES (112,'DEVELOPER TOOLS',103);
INSERT INTO COMPANY VALUES (113,'WINDOWS',103);
INSERT INTO COMPANY VALUES (114,'ENTERTAINMENT',103);
INSERT INTO COMPANY VALUES (115,'GAMES',114);
INSERT INTO COMPANY VALUES (116,'MULTIMEDIA',114);
INSERT INTO COMPANY VALUES (117,'EDUCATION',101);
COMMIT;

The first value in each record is the node number, and the third value is the recursively defined parent
node number.

In order to build the required path table, you can invoke an SQL common table expression (CTE) with
an embedded UNION ALL statement. This specific combination of the CTE together with UNION ALL
allows the first SELECT statement in the UNION ALL to act as the anchor query and the second SELECT
statement to act as the recursive query. For this situation, ANSI standard SQL also exposes the
pseudo-column level which counts the number of levels down from the topmost node. For more
information on this advanced SQL syntax, theres a Wikipedia article entitled Hierarchical and
Recursive Queries in SQL.

Using CTE with UNION ALL, you can build the desired hierarchy bridge table with the SELECT statement
following the CTE in the following:

Kimball Group. All rights reserved.